JP2007220049A - Image-based protruded displacement mapping method and bi-layered displacement mapping method using the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method of representing the details of the surface of an object, which allows an object surface to protrude from a reference surface representing an object, without modifying the reference surface or adding micro-polygons to the reference surface. <P>SOLUTION: This image-based protruded displacement mapping method includes: intersection point search steps 401, 402 of searching for an intersection point of the surface of a protruding object, defined by displacement information and an observer's sight line and displaying the shape of the protruding object on the polygonal surface; an overflow representation step 403 of adding a separate polygonal surface and representing a shape of the protruding object corresponding to overflow deviating from a boundary of the polygonal surface, on the added polygonal surface; a steeply inclined surface detection step 404 of detecting a steeply inclined surface in the shape of the protruding object; and a steeply inclined surface correction step 404 of representing an additional texture on the steeply inclined surface. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention is a technology applied to an entertainment information device such as a video game machine and a computer game, an information device expressing terrain space information, or an information device performing three-dimensional rendering using computer graphics. The present invention relates to a projecting displacement mapping method that expresses so as to project from a surface. The present invention also relates to a double displacement mapping method using a mixture of the image-based protruding displacement mapping technology and a general shape-based displacement mapping technology.

  In the computer graphics field, a method for expressing details of an object surface from the concept of bump mapping proposed by Blinn (Non-patent Document 1) and displacement mapping proposed by Cook (Non-patent Document 2) has been proposed.

Conventionally proposed object surface detailed expression methods are roughly classified into a shape-based triangular network generation method and a video-based displacement mapping method.
The triangular mesh generation method is a method of generating details of the object surface by generating a large number of fine polygons. This method generates a large number of fine polygons and is difficult to apply to interactive programs. is there. In order to solve such a problem, a method of expressing the displacement of the object surface by explicitly rendering with fine polygons has been proposed. Proposed methods include a ray projection method, a three-dimensional backward image warping method, a three-dimensional texture mapping method, and a method for calculating visible information in advance. Among these methods, the light projection method and the reverse image warping method have a problem that they are not suitable for real-time applications because of the large amount of calculation. The three-dimensional texture mapping method has a problem that an unnatural result can be obtained depending on the viewpoint because the displacement mapping is rendered by layering polygons having a two-dimensional texture. In the method of calculating the visible information in advance, there is a problem that a large amount of memory must be used to store the sample of the five-dimensional function.

  The displacement mapping method based on an image is a method of expressing the surface details of an object without generating fine polygons, and uses a technique that simplifies or replicates ray-tracing. However, such a conventional method does not commonly project the object surface on the polygonal surface, but uses a method in which the bottom surface of the object is placed under the polygonal surface to reduce the detailed shape of the object surface. Express.

  FIG. 1 is a diagram showing the concept of a conventional displacement mapping method based on video. In FIG. 1, reference numeral 101 is an actual polygon surface, 102 is the depth from the actual polygon surface to the surface of the object, 103 is the height from the bottom surface to the surface of the object, and 104 is the viewpoint.

  Conventionally, since the polygonal surface is positioned at the uppermost surface in a virtual space composed of the displacement map of the object, it actually represents the depth, not the height of the object. Therefore, the conventional method is suitable for expressing a shape deviated from the polygon surface which is the reference surface, but has various problems when expressing a shape protruding from the polygon surface.

  FIG. 2 is a diagram for explaining a problem of the conventional image-based displacement mapping method, in which (a) shows a state in which a texture that does not reflect a polygon surface and a displacement map is rendered, and (b) and ( c) shows a state in which a texture reflecting the polygon surface and the displacement map is rendered, and (d) shows a state in which the texture reflecting the polygon surface and the displacement map and its periphery are processed.

As shown in (a), when the texture (T) that does not reflect the polygon plane (P) and the displacement map is rendered, the height of the object is not reflected.
When the displacement map is reflected on the texture in the diagram (a) using the conventional image-based displacement mapping method, the bottom surface is turned down as shown in (b) and (c), thereby A region (B) in which texture information leaks occurs around the boundary. If the height of the object is to be expressed higher, the bottom surface of the object must be made deeper than the polygon surface, so that such an information leakage area (B) becomes wider as the height of the object increases. (The information leakage area in (c) is wider than the information leakage area in (b)).

  The problems of the above-described conventional displacement mapping method based on video can be summarized as follows. Since the texture is not projected on the polygonal surface but is made to appear downward in the polygonal surface, the actual detail position changes depending on the surface of the bottom surface. In addition, the bottom surface becomes farther away from the human viewpoint while leaning down, causing a phenomenon that the bottom surface contracts.

Due to the above-described problems, in the conventional displacement mapping method based on an image, the same tile-shaped shape is repeatedly arranged as shown in FIG. When the height is not high, details of the object surface can be expressed, but in other cases, details of the object surface cannot be expressed accurately.
"Simulation of Wrinkled Surfaces", Proceedings of International Conference on Computer Graphics and Interactive Technologies, pp. 286-292, 1978. “Shade Trees”, Proceedings of International Conference on Computer Graphics and Interactive Technologies, pp. 223-231, 1984.

  The present invention has been made to solve the above-described problems of the prior art, and an object of the present invention is to deform a reference surface (polygon surface) representing an object or to change the reference surface to the reference surface. An object of the present invention is to provide a method for expressing details of an object surface so that a portion protruding from a reference surface has a shape protruding from the reference surface without adding a fine polygon.

Another object of the present invention is to provide a method for processing an excess flow that falls outside a polygonal surface when the height of the texture is projected from the reference surface.
Still another object of the present invention is to use a shape-based displacement mapping method for expressing details of surface of terrain and a video-based protruding displacement mapping method for expressing details of buildings protruding on the terrain. It is to provide a double displacement mapping method.

  The image-based projecting displacement mapping method according to the present invention for achieving the above-mentioned object uses an image-based projecting displacement in a system that expresses an object shape projecting from a polygon surface using a three-dimensional object, texture information, and displacement information. In the mapping method, an intersection search step of searching for an intersection between the surface of the protruding object defined by the displacement information and the observer's line of sight and displaying the shape of the protruding object on the polygon surface, and an additional polygon surface In addition, an excess display step for displaying the shape of the protruding object corresponding to the excess exceeding the boundary of the polygon surface on the additional polygon surface, and a steeply inclined surface in the shape of the protruding object are detected. A steeply inclined surface detecting step and a steeply inclined surface correcting step of expressing an additional texture on the steeply inclined surface. To.

  Further, the double displacement mapping method according to the present invention is the double displacement mapping method in the system for representing the shape of the object using the three-dimensional object, the texture information, the global displacement information, and the regional displacement information. And a global displacement mapping stage that represents the surface of the object based on polygons using the global displacement information, and a regional displacement that represents a detailed shape of the surface of the object using the texture information and the regional displacement information. Mapping stage, occlusion resolution stage for arranging the objects subjected to the global displacement mapping and the regional displacement mapping in order of distance from the viewpoint, and the height direction of the tangent space between adjacent polygons in the global displacement mapping stage And the same tangent space correction stage.

  According to another aspect of the present invention, there is provided a method for processing an excess exceeding a boundary of a polygon surface by a computer graphic system that displays a shape of an object protruding on the polygon surface. A vertical plane of the polygon plane is added adjacent to the boundary of the generated polygon plane, and the protruding object shape corresponding to the excess is displayed on the added vertical plane.

  As described above, according to the present invention, it is possible to accurately represent a silhouette having a shape that actually protrudes from a polygon surface. Unlike the conventional image-based displacement mapping method using the GPU, which can only represent the shape depressed from the reference surface, according to the present invention, the displacement map is represented so as to actually represent the shape protruding from the polygon surface. Can be used to express a silhouette with a protruding shape. Further, by simply adding two auxiliary surfaces without deforming the shape of the polygon surface, the details of the protruding shape can be expressed with almost no increase in the number of vertices of the overall model.

  Furthermore, according to the projecting displacement mapping method of the present invention, it is possible to express a projecting shape having a steep inclination by adding details of the inclined surface in the calculation process in units of pixels. The details of the steep slope do not affect the rendering performance because of the use of very small tiled textures.

  Further, the projecting displacement mapping method of the present invention can be applied to a vertically projecting shape such as a high-rise building, and can be used in parallel with a graphic algorithm based on a poly model, It can be used to improve the representation of topographic space information. Thus, in the video representation of entertainment information devices such as video games and computer games and information devices using computer graphics, it can be used to add detail to the surface of the object and improve the quality of the video.

  The present invention has an advantage that a protruding shape and a silhouette of the shape can be expressed without adding a fine polygon to the surface of the object.

Hereinafter, an image-based protruding displacement mapping method according to the present invention and a double displacement mapping method using the method will be described in detail with reference to the accompanying drawings.
The present invention proposes a method for expressing details of a shape protruding from a reference surface without deforming or distorting the reference surface (polygon surface) expressing the object. In the present invention, the displacement map of the surface protruding from the reference plane is defined, and thereby the protrusion is expressed using a ray-tracer. To solve the excess displacement that occurs at the boundary of the protruding object, two additional surfaces are added to the reference surface to represent the silhouette.

Prior to describing the present invention, terms used in the present invention will be explained.
-Rendering: The meaning commonly used in the field of computer graphics refers to the process of obtaining a two-dimensional image from a three-dimensional model through a series of graphics pipelines. There are two methods for obtaining 2D images such as photographs from various scene information showing objects, lighting, texture, position, hue, etc. There are a variety of ways to get it. A calculation process for obtaining a two-dimensional image or a moving image from a three-dimensional model is called rendering.

  -Representing surface details: When representing an object through a rendering process from geometry information (generally geometry information means a polygon model), which is information on the shape of the 3D object, the surface texture of the object, This refers to expressing textures of fine shapes that do not appear in geometry information, shadows due to lighting, and fine shapes that are projected or depressed. That is, it expresses a visual texture such as skin wrinkles or wood grain and bends such as wrinkles on the surface of a smooth model where polygons appear.

  -Representing the displacement of the surface: In expressing the details of the surface, in particular, it means expressing fine changes in height, which are caused by wrinkles, wood grain, etc. from the surface of the object composed of polygons. For example, when the shape of an animal is modeled with a smooth polygon, this means expressing the heights of wrinkles and protruding protrusions that are squeezed toward the inside of the skin by adjusting the fineness of the surface.

  -Displacement mapping: A method of expressing surface displacement. When rendering a polygon representing an object using a displacement map that records the displacement in the normal direction of the surface of the object, the displacement is displayed. The surface displacement is expressed by the displacement of each pixel recorded in the map.

  -Displacement map: Refers to a mapping table or raster data in which displacement information on the surface of an object is stored as a raster value. Of these displacement maps, a representation of height information is called a height map.

  FIG. 3 is a view for explaining the concept of the image-based protrusion displacement mapping method according to the present invention. In the present invention, a method for expressing a protruding displacement for expressing a shape protruding from a polygon surface which is a reference surface is proposed. In FIG. 3, reference numeral 301 denotes a polygon surface which is a reference surface, 302 denotes a depth from the reference surface to the object surface, 303 denotes a height from the reference surface to the object surface, and 304 denotes a user's viewpoint. is there.

A displacement map having a shape protruding from the surface is defined with 0 representing the height of the polygon surface as a reference surface and 1 representing the highest position.
In FIG. 3, the pixel to be rendered is a point A ′ on the polygon surface, and the point to be actually shown is a point P that protrudes depending on the height. Therefore, in order to render the point A ′ on the polygonal surface, a point P ′ where the line of sight A′-B ′ and the surface of the object intersect must be searched. When the line of sight A′-B ′ intersects the surface of the object many times, the intersection of the highest displacement is searched and that point is set to P ′.

  That is, in the prior art, the point that should actually be shown at the point A on the polygonal surface that is the reference surface is the depressed point P, but in the present invention, the point that should actually be shown at the point A ′ on the polygonal surface is protruding. P ′.

FIG. 4 is a diagram illustrating a process of an image-based protrusion displacement mapping method according to the present invention.
This image-based protrusion displacement mapping method can be applied to a general computer system. The video-based projecting displacement mapping system 400 receives three-dimensional object information from an object database (DB) 410 and receives surface texture information and regional displacement information from a surface texture database 420 and a displacement map database 430, respectively.

  The image-based projecting displacement mapping method of the system 400 includes a sequential search stage (401), a binary search stage (402), an excess processing stage (403), and an abrupt slope detection and correction stage (404). Become. This image-based protruding displacement mapping method can express abrupt displacement and details of an object surface without adding fine polygons by using a displacement map and surface texture information.

Hereinafter, each step of the image-based protrusion displacement mapping method will be described in detail.
Sequential search stage (401)
This sequential search step is a process of searching for the approximate position of the intersection between the surface of the object and the line of sight defined by the displacement map, which will be described with reference to FIG.

  As shown in FIG. 5, the upper surface and the reference surface of the virtual volume formed by the displacement map are divided at regular intervals. For the points B and 1-4 where the upper surface and the divided surface of the virtual volume intersect with the line of sight connecting the viewpoint 304 and the point A to be rendered, the height values and the displacement map gradually divided from the upper surface are stored. Compare the measured height value. If the divided height value is higher than the height value stored in the displacement map, continue the inspection for the next point, and if the divided height value is lower than the height value stored in the displacement map, Stop. For example, in the case of FIG. 5, when the height value of point B, which is the intersection of the upper surface and the viewpoint, is compared with the height value stored in the displacement map at point B, the height value of point B is higher. The inspection for the next intersection 1 is continued. Since the height value of point 1 is higher than the height value stored in the displacement map at that point 1, the inspection for the next intersection 2 continues. After the inspection up to the intersection 3 in this way, the inspection at the next intersection 4 is completed because the height value of the intersection 4 is lower than the height value stored in the displacement map at the point 4.

  Through this sequential search step, the approximate position of the intersection of the surface of the object and the line of sight defined by the displacement map can be found. In the case of FIG. 5, it can be seen that the intersection is located between 3 and 4.

Binary search stage (402)
After obtaining the approximate position of the intersection between the surface of the object defined by the displacement map and the line of sight by the sequential search step as described above, a binary search step as shown in FIG. 6 is performed to obtain a more accurate intersection point. Get information on height h. FIG. 6 is an enlarged view of the shape between the points 3 and 4 in FIG. 5, and the information on the accurate height h of the intersection is searched by performing a binary search between the points 3 and 4. FIG.

  The binary search method performed in the binary search stage is a paper co-authored by Polarpo, Oliveira, Comba [“Real-time Relief Mapping on Arbitrary Polygonal Surfaces”, Proceedings of Symposium on Intersci- encePop. 155-162, 2005. ] In detail.

  When the intersection between the line of sight and the surface of the object is obtained in this way, the texture color value at the intersection is read from the surface texture database 420 in FIG. 4 and displayed on the corresponding pixel (point A) on the polygon surface 301.

Excess processing stage (403)
FIG. 7 is a diagram illustrating a process of rendering a quadrilateral polygon three-dimensionally by applying the image-based protruding displacement mapping method according to the present invention.

  FIG. 7A shows the result of rendering a texture that does not reflect the polygon plane and the displacement map. The sequential search step and the binary search step according to the present invention are performed to obtain the intersection between the surface of the object defined by the displacement map and the line of sight, and the texture color value corresponding to the intersection is read and displayed on the pixel of the polygon surface. By repeating the process, as shown in FIG. 7B, the result of rendering the texture reflecting the polygon plane and the displacement map is obtained.

  When (b) in FIG. 7 is compared with (b) and (c) in FIG. 2, in the rendering results in (b) and (c) in FIG. In the rendering result of FIG. 7B, the bottom surface has a height of 0 and is not deformed and coincides with the polygon surface. The shape is such that the object surface protrudes above the polygon surface by protruding only in the form of a box on the polygon surface. It has become.

  Further, as shown in FIG. 7C, it is confirmed that even if the texture height is doubled, the bottom surface is not deformed in conformity with the polygonal surface and is doubled by the height of the box. be able to. This can be contrasted with the case in FIG. 2C where the bottom surface distortion becomes worse as the wrinkle depth becomes deeper.

  Since the projection displacement mapping of the present invention is performed in units of pixels, if the shape protruding from the surface is expressed by displacement mapping, the region protruding above the line of sight EV exceeds the polygonal surface as shown in FIG. It is impossible to express.

  For example, in order to represent the point P by the displacement mapping in the protruding portion on the upper side of the line of sight EV in FIG. 9, it must be represented at the point A exceeding the boundary of the polygonal surface EF. However, displacement mapping calculation in units of pixels is not performed in an area outside the boundary of the actual polygon surface. Such a problem due to the excess of the pixel area is a problem that commonly occurs in the video-based displacement mapping technology, and is also the reason why the outstanding details of the conventional technology are expressed by the 陷 沒 method.

  Therefore, if the problem due to the excess of pixels due to such displacement cannot be overcome, the contour line protruding in the vicinity of the boundary of the polygon surface is accurately expressed as shown in FIG. The problem of being unable to do so occurs.

  In the present invention, two methods for solving the problem due to excess are proposed. The first method is a method of expanding the original polygon surface EF, and the second method is a method of adding a surface perpendicular to the original polygon surface.

Method for Extending Original Polygon Surface Since point P in FIG. 9 is above line segment EV connecting end point E of polygon surface and viewpoint V, displacement mapping of this point is impossible. However, if the polygon surface EF in FIG. 9 is expanded to ED ′, rendering of the point P at the point A of the expanded polygon surface 901 becomes possible.

  In the present invention, as shown in FIGS. 9 and 10, the polygon surface is expanded in order to express the protruding shape. In order to express all the protruding shapes by the displacement map, the polygonal surface must be extended to the intersection D ′ where the line VD connecting the viewpoint V and the point D intersects with the extended surface of the polygonal surface EF. Therefore, the polygon surface EF is expanded to D′-F.

  However, the closer the inclination of the line of sight VA when viewing the point P in FIG. 9 is to the inclination of the polygon surface EF, the farther the position of the point A is, the closer to infinity, D ′. Therefore, the lengths Ex and Ey of the extended polygon surface 901 displayed in FIG. 10 are also close to infinity. There is a problem that the resolution of the texture is lowered due to the tilt of the viewpoint and the image quality is lowered.

  As shown in FIG. 10, the method of expanding the polygonal surface has an advantage of not increasing the number of vertices, but the size to be expanded must be calculated according to the change of the viewpoint. There is a disadvantage that the quality is lowered. In addition, since the calculation in units of vertices and the calculation in units of pixels must all be performed, there are restrictions on implementation on a GPU (graphic process unit).

In the present invention, a method of generating an additional surface is used to solve such a problem.
Method for Adding a Vertical Surface FIG. 11 is an expanded representation of FIG. 9 in three dimensions. The line of sight for the point P in FIG. 9 intersects the plane 902 perpendicular to the polygon plane EF at the point C. Therefore, as shown in FIG. 9, if a vertical plane 902 perpendicular to the polygon plane is added and the texture information of the point P is rendered on the pixel C of the vertical plane 902, the problem of excess area can be solved. That is, if the point C is mapped to the point A existing on the virtual extended polygon plane 901E-D 'and the texture coordinates of this point are used, it exists on the vertical surface using only one displacement map and texture. The point C to be expressed can be expressed.

  The ratio (vx, vy, vz) of the component of the line of sight AV in FIG. 11 can be obtained from the coordinates of the viewpoint in the tangent space. The uv space coordinates of the point C on the vertical plane 902 are expressed by (dx, dy). If the trigonometric ratio is used from these relationships, the uv space coordinates of the point A (x ′, y ′) mapped to the virtual expansion plane 903 can be obtained as in Equation 1 below.

Here, (vx, vy, vz) is a value in the tangent space coordinate system of the polygon surface 301, (dx, dy) is a value in the uv space coordinate system of the vertical surface 902, and (x ', y') Is a value in the uv space coordinate system of the virtual extension plane 903. The uv space of the virtual extended surface 903 is defined as the uv space warped with reference to the boundary of the polygon surface 301.

  The tangent space coordinates of the vertical surface 902 in FIG. 11 can be calculated using the relationship of the tangent space coordinate system between the polygon surface and the adjacent surface as shown in FIG. When the bottom surface of FIG. 12 is a polygon surface, the tangent space of the surface perpendicular to the left side of the polygon surface is converted as shown in the following equation (2).

Here, normal is a normal vector, binoral is a vertical normal vector, and tangent is a tangent vector.

If the definition of the tangent space coordinate system in FIG. 12 is used, the tangent space can be mapped to surfaces adjacent in other directions.
In the present invention, as shown in FIG. 13, a polygonal vertical surface 902 is added so as to be in contact with the boundary where the region excess occurs depending on the viewpoint direction. Then, by expressing the protruding displacement of the object on the added vertical plane, it is possible to represent the silhouette for the excess exceeding the polygon plane area.

  The added vertical surface rendering uses alpha blending to selectively visualize only the protruding areas of the object. The method presented in the present invention can be expressed by one texture and displacement map by mapping the tangent space. Therefore, the silhouette of the object can be expressed with a small amount of memory usage, and an accurate boundary can be expressed.

  As shown in FIG. 7C, in the situation where the outline of the shape protruding near the boundary of the polygon surface is not accurately expressed, one vertical surface of the polygon surface is added to the boundary of the polygon surface where the excess occurs. In addition, the result of rendering the excess is shown in FIG. 7D, and the result of rendering the excess by adding two vertical surfaces of the polygon surface to the boundary of the polygon surface where the excess occurs is shown in FIG. ). From FIG. 7E, it can be seen that a complete silhouette is expressed near the boundary of the projecting displacement mapping.

Rapid slope detection and correction stage (404)
The displacement mapping method for expressing the protruding shape presented in the present invention can also be applied to the expression of an object with a steep inclination of the protruding shape, particularly an object protruding in the vertical direction like a building. Since an object protruding vertically has a very steep inclination in the protruding direction, a point on a texture coordinate may represent a wide area depending on the height. Therefore, good video quality cannot be obtained only with the existing method.

  The soft displacement map is expressed softly because the size of the texture coordinates deformed by the displacement is not significantly different from the neighboring pixels, whereas the vertical plane of the object protruding vertically is one on the displacement map. Since it is expressed as a point or a line, the distortion of the texture becomes very large depending on the viewpoint. Therefore, there is a need for a method that can compensate for the sudden displacement of the projected displacement.

  If the protruding shape with a steep slope is exaggerated, it can be assumed that it is a hexahedral shape protruding vertically as shown in FIG. If this vertically projecting hexahedron shape can be accurately expressed, it is possible to express a protrusion with respect to any steeply inclined surface.

  FIG. 7F shows the result of adding the side surface of the box as additional texture information in the result of FIG. In this case, since the side surface of the box is perpendicular to the polygon surface, one pixel represents all of the pixel from the lowest point to the highest point on the texture coordinate. Therefore, as shown in FIG. 7 (e), it is expressed as a shape in which the texture is extended to the side. Although the boundary line between the base 701 and the texture 702 in FIG. 7A represents the entire side surface of the shape protruding in FIG. 7E, a fine error occurs in the sequential search and binary search processes in pixel units. Therefore, an effect appears in which pixels around the black line corresponding to the boundary line are linearly interpolated over the entire side of the box.

  If the protruding shape is divided into the side of the box (steeply inclined surface) and the upper surface (gradually inclined surface), additional texture information is applied to the steeply inclined surface where the distortion is severely generated, and it occurs on the side of the box. Distortion can be corrected.

For this reason, it must be possible to distinguish a sharply inclined surface and a gently inclined surface having a shape that protrudes first.
Detection of an abrupt inclined surface is possible by the following method. FIG. 14 is similar to the cross section of the displacement map used in FIG. In the case of a point P _A existing on a gently inclined surface, a displacement map on the uv coordinate of the first point is read in the sequential search process, and h _A1 is detected with the height value of P _A1 . In the next stage, since the height value of the second point is lower than the value read from the displacement map, the search is interrupted sequentially. Thereafter, the exact intersection obtained by the binary search is P _A , and the height of this point is determined as h _A. At this time, the height h _A of the point P _A existing on the gentle slope has a value substantially equal to h _A1 .

On the other hand, in the case of the point P _B located on the steeply inclined surface, the following result is obtained. In the sequential search process, the value h _B1 read from the displacement map up to the second position is located on the bottom surface. In the next stage, the height value of the third point is smaller than the value read from the displacement map (same as h _A1 ), and the search is sequentially interrupted. Thereafter, the height value of the intersection point P _B obtained by the binary search is determined as h _B. In this case, the height h _B of the P _B points located abrupt inclined surface significantly different from h _B1.

Thus, the point located on the steeply inclined surface can be classified using the difference between the two cases. The classification process is as follows.
1. The displacement map value h _n−1 read at the coordinates immediately before the end of detection in the sequential search process is stored.

2. The height value h of the intersection obtained from the binary search result is obtained.
3. If the difference h _n−1 −h between the two height values is larger than the limit value, it is determined that the point is located on a surface having a steep slope. If the difference between the two height values is smaller than the limit value, it is determined that the point is located on a surface having a gentle slope.

  A tile-shaped texture that occupies a small amount of memory can be mapped to the steeply inclined surface detected by such a method using the uv coordinates and the detected height value. At this time, if the texture coordinate system of the tile form is an (x, y) two-dimensional coordinate, the following equation 3 can be used to add details of the inclined surface with respect to a sudden displacement.

Here, u and v are (u, v) coordinate values of each pixel existing on the polygon surface. h is the height value of the intersection obtained in the protrusion displacement mapping process.

Color information is sampled from the tile-shaped texture using the x and y values calculated by Equation 3, and used as the color value of each pixel on the steeply inclined surface.
However, Equation 3 is only an example of additional texture mapping for a sudden displacement, and in particular, d is appropriately defined according to the shape of the object and defined as a function of u and v. As the simplest example, d = u + v can be used as a linear combination of u and v. Hereinafter, the process of obtaining color information from the obtained x and y values is similar to a general texture sampling process.

The result of covering the steeply inclined surface with the tile-shaped texture is shown in FIG.
This method can be used to express an object protruding in the vertical direction, such as a building.
At this time, the limit value that determines the steeply inclined surface and the gently inclined surface must be appropriately adjusted according to the resolution of the screen and the displacement map. If the limit value is set too small, there is a problem that pixels that are detected as an inclined surface are mixed because pixels are detected as a sharply inclined surface up to a flat surface. Such a problem becomes more serious when there is a minute error in the value of the displacement map. In this case, the error is removed by adjusting the limit value.

Double Displacement Mapping Method FIG. 15 illustrates a system for double displacement mapping according to the present invention.
The double displacement mapping system 1500 of the present invention comprises a polygon-based global displacement mapping means 1501 and a regional displacement mapping means 1502 for expressing details of the polygon surface. The global displacement mapping means 1501 is provided with global displacement information from the global displacement map DM1, and the regional displacement mapping means 1502 is provided with regional displacement information from the regional displacement map DM2. Further, texture information (TM) is provided to the regional displacement mapping means 1502.

  The double displacement mapping system 1500 of the present invention receives information of an object to be rendered from an object database 1510 of a computer graphic system, and receives a global displacement map DM1 and a regional displacement map DM2 and texture information TM used for displacement mapping from a memory 1520. Receiving and rendering the surface of the object in more detail to generate an image 1530 in real time.

  The global displacement mapping means 1501 is used to represent the soft displacement of the object. The details of the surface are expressed by adjusting the heights of the vertices constituting the surface of the object from the global displacement map DM1 of FIG. At this time, polygon-based displacement mapping is used, and the number of vertices can be increased for more precise representation.

  The regional displacement mapping unit 1502 is used to express a more detailed displacement that the global displacement mapping unit 1501 could not express. The local displacement mapping means 1502 is a displacement mapping method for expressing a shape that protrudes or protrudes on the polygon surface, and the image-based protruding displacement mapping method proposed in FIG. 4 of the present invention is applied to the protruding shape. Thus, the rapid displacement and detailed shape of the object surface are expressed without adding fine polygons.

The regional displacement mapping means of the present invention executes the protruding displacement mapping method for the protruding shape.
When an image is displayed on the screen after executing the polygon-based global displacement mapping and the polygon surface regional displacement mapping, the occlusion solving process (1503) of the protruding shape and the tangent space correcting process (1504) are executed. .

First, the protruding occlusion solution process (1503) will be described.
The occlusion solution of the protruding shape is executed by setting the displacement mapping of the polygon surface far from the viewpoint to occur first, and setting the displacement mapping of the polygon surface close to the viewpoint to occur later. As a result, the polygonal surface that is far from the viewpoint is blocked by the polygonal surface close to the viewpoint.

Next, the necessity and method of the tangent space correction process (1504) will be described.
If projecting displacement mapping is applied on a single flat polygon, the displacement map is located on a plane having the same normal line, so that all pixels have the same tangent space. However, if projecting displacement mapping is applied to a curved surface to which global displacement mapping having a grid or TIN (triangular irregular networks) form is applied for double displacement mapping, the difference in normal direction between adjacent polygons As a result, an error occurs in the protruding shape. In order to solve such a problem, the tangent space is corrected in the double displacement mapping. This tangent space correction is performed by making the height direction of the tangent space between adjacent polygons the same.

  The occlusion solution process and the tangent space correction process described above may be included in the double displacement mapping method as described above, or may be included in the protruding displacement mapping process of FIG.

Embodiments The present invention can be applied to entertainment information devices including video games and computer games, information devices including representations of topographical space information, and information devices that generate images by computer graphic rendering.

  In particular, when the present invention is applied to modeling and visualization of actual terrain space information, the configuration is as shown in FIG. In order to express an urban terrain model using displacement mapping, it is necessary to be able to express all of the relatively large displacements of the basic terrain and the relatively small displacements generated by the features. Due to the characteristics of geographic information, when the displacement of the basic terrain and the displacement of a feature such as a building are represented at the same time, it is difficult to represent the exact shape of the building. It is advantageous in terms of supply and demand and visualization of geographic information to separate building displacement from basic topographic displacement. Therefore, after the global displacement mapping using the basic topography and the regional displacement mapping using the building height are executed independently, the two displacements are merged.

  A displacement mapping method suitable for each application is required for the global displacement mapping and the regional displacement mapping constituting the double displacement mapping. In terms of modeling and visualization, it is reasonable to use methods that are appropriate for two-stage displacement mapping. In the present invention, the shape-based global displacement mapping method is used to represent the topography, and the image-based projected displacement mapping method is used in a mixed manner to represent the protruding building.

  As shown in FIG. 16, for the modeling and expression of the urban virtual environment, it is classified into a global displacement mapping 1501 for the representation of the basic terrain and a regional displacement mapping 1502 for the representation of the features such as buildings. Displacement mapping can be applied. The basic terrain expressed by the height of a pure index excluding artificial terrain features such as buildings changes insensitively over time, while buildings continuously change relatively quickly according to changes in the urban environment. The building displacement map 1611 required for regional displacement mapping must be updated more frequently. Therefore, the building displacement map 1611 that stores the artificial terrain features and the digital terrain model 1621 that stores the basic terrain are separated to easily cope with changes in the urban environment. That is, by updating only the building displacement map 1611 generated from the numerical map 1622, it is possible to easily update the urban virtual environment.

  The double displacement mapping method of FIG. 16 uses the method of expressing the height of the terrain by the global displacement mapping 1501 and the height of the building shape by the regional displacement mapping 1502 on the basic terrain. Separate easy-to-do buildings from difficult-to-change basic terrain. The global displacement mapping 1501 expresses the displacement of the basic terrain from the digital terrain model 1621, and on this, the texture information 1612 using the building displacement map 1611 and the aerial photograph 1623 generated from the numerical map 1622 is used. A regional displacement mapping 1502 is performed that represents the protruding details.

  As a result of applying the double displacement mapping and the protruding displacement mapping described in the present invention to the expression of actual topographical space information as shown in FIG. 16, an image as shown in FIG. 17 is generated in real time.

  The present invention is applied to a projecting displacement mapping method for expressing an object as projecting from a surface, and a double displacement mapping method using a mixture of this image-based projecting displacement mapping technology and a general shape-based displacement mapping technology. Is possible.

It is a figure which shows the concept of the displacement mapping method based on the conventional image | video. It is a figure shown in order to demonstrate the problem of the displacement mapping method based on the conventional image | video. FIG. 5 is a diagram illustrating a concept of a video-based protruding displacement mapping method according to the present invention. FIG. 5 is a diagram illustrating a process of a video base protrusion displacement mapping method according to the present invention. It is a figure which shows the process of searching for the approximate position of the intersection of the surface of the object defined by the displacement map and the line of sight in the sequential search stage. It is a figure which shows the process of searching for the exact intersection of the surface of an object and a gaze in a binary search stage. FIG. 5 is a diagram illustrating a process of rendering a quadrilateral polygon three-dimensionally by applying the image-based protruding displacement mapping method according to the present invention. FIG. 5 is a diagram illustrating a problem due to an excess of pixel displacement that is generally generated in a video base protrusion displacement mapping process. FIG. 9 is a diagram simply illustrating a process of solving the displacement excess problem of FIG. 8. It is a figure which shows the state which expanded the polygon surface in order to solve the excess part problem. It is a figure which expands and shows the solution method of the excess part problem in three-dimensional space. It is a figure which shows the mapping relationship of a contact space. It is a figure which shows the state which added the surface perpendicular | vertical to the polygonal surface in order to solve the excess amount problem. It is a figure which shows the judgment process of a sudden displacement and a gentle displacement. 1 is a functional block diagram of a system for performing double displacement mapping according to the present invention. FIG. It is a figure which shows the example which utilizes the double displacement mapping and protrusion displacement mapping which were demonstrated by this invention for the expression of topographical space information. It is a figure which shows the image | video produced | generated in real time from the Example of FIG.

Explanation of symbols

304 ... Viewpoint, 400 ... System for video-based protruding displacement mapping, 401 ... Sequential search stage, 402 ... Binary search stage, 403 ... Excess processing stage, 404 ... Abrupt slope detection and correction stage, 410 ... Object database , 420 ... surface texture database, 430 ... displacement map database, 701 ... base, 702 ... texture, 901 ... polygon surface, 902 ... surface perpendicular to the polygon surface, 903 ... virtual extension surface, 1500 ... double displacement mapping System, 1501 ... Global displacement mapping means, 1502 ... Regional displacement mapping means, 1510 ... Object database, 1520 ... Memory, 1530 ... Image, 1611 ... Building displacement map, 1612 ... Texture information, 1621 ... Digital terrain model, 1622 ... Number Map, 1623 ... aerial photographs

Claims (27)

In a video-based projecting displacement mapping method in a system that expresses an object shape projecting from a polygon surface using a 3D object, texture information, and displacement information,
Searching for the intersection of the surface of the protruding object defined by the displacement information and the observer's line of sight, and displaying the shape of the protruding object on the polygon surface; and
Adding an additional polygon surface, and displaying the shape of the protruding object corresponding to the excess exceeding the boundary of the polygon surface on the additional polygon surface; and
A steep slope detection step of detecting a steep slope in the shape of the protruding object;
A steep slope correction step for expressing an additional texture on the steep slope;
An image-based projecting displacement mapping method comprising: The intersection search step includes:
A sequential search step for searching for an approximate position of an intersection between the surface of the object defined by the displacement information and the line of sight;
A binary search step for searching for an exact intersection point and a height of the intersection point from a rough position of the intersection point;
The image-based projecting displacement mapping method according to claim 1, further comprising: The sequential search step includes:
Dividing the upper surface of the virtual volume constituted by the displacement information and the polygon surface at a constant interval,
Search for a divided intersection where the upper surface of the virtual volume and the divided surface and the line of sight intersect,
Starting with the uppermost divided intersection, the height value of the divided surface is compared with the height value stored in the displacement information, and the height value of the divided surface is stored in the displacement information. If the height value is higher than the height value stored in the displacement information, the inspection is continued if the height value of the divided surface is lower than the height value stored in the displacement information.
3. The image-based projecting displacement mapping method according to claim 2, wherein an approximate position of the intersection is set between the divided intersection where the inspection is stopped and the divided intersection on the intersection. The excess display step includes:
The extended polygon surface is expanded on the same surface as the additional polygon surface, and the shape of the protruding object corresponding to the excess is displayed on the expanded polygon surface. The image base protrusion displacement mapping method described. The excess display step includes:
Depending on the direction of the line of sight, a polygonal surface perpendicular to the boundary of the polygonal surface where excess occurs is added as the additional polygonal surface, and the added vertical surface corresponds to the excess. The method of claim 1, wherein the shape of the protruding object is displayed.   The method of claim 5, wherein the excess display step maps the tangent space between the polygon surface and the vertical surface. The excess display step includes:
The spatial coordinates of the intersection C between the line of sight and the vertical plane are (dx, dy), and the spatial coordinates of the intersection A between the virtual extension plane obtained by extending the polygon plane on the same plane and the line of sight are (x ′, y '), The point C is mapped to the point A using the following formula:
Here, (vx, vy, vz) is a value in the tangent space coordinate system of the polygon surface, (dx, dy) is a value in the uv space coordinate system of the vertical surface, and (x ′, y ′) 6. The image-based protruding displacement mapping method according to claim 5, wherein is a value of a spatial coordinate system of the virtual extension plane, and the shape of the protruding object is displayed using the texture coordinates of the point A. .   The image-based projecting displacement mapping according to claim 1, wherein the excess display step does not display the additional polygon surface, but only renders the shape of the protruding object corresponding to the excess. Method. The steeply inclined surface detecting step includes
Dividing the upper surface of the virtual volume constituted by the displacement information and the polygon surface at a constant interval,
Search for a divided intersection where the upper surface of the virtual volume and the divided surface and the line of sight intersect,
Obtaining a height value of an intersection between the surface of the object defined by the displacement information and the line of sight, and setting the obtained height value as a first height value;
Obtaining a height value of a divided intersection existing on the upper surface adjacent to the intersection, and setting the obtained height value as a second height value;
The image base according to claim 1, wherein when the difference between the first height value and the second height value exceeds a limit value, the image base is determined to be a steeply inclined surface. Projection displacement mapping method.   The method of claim 1, wherein the steep slope correction step adds a tile-shaped texture to the steep slope.   The method of claim 1, further comprising an occlusion solving step of arranging a number of objects that have undergone the steep slope correction step in order of distance from a viewpoint. In a double displacement mapping method in a system that represents a shape of an object using a three-dimensional object, texture information, global displacement information, and regional displacement information,
A global displacement mapping step of representing the surface of the object based on polygons using the 3D object and global displacement information;
A regional displacement mapping stage that represents a detailed shape of the surface of the object using the texture information and the regional displacement information;
An occlusion solving step of arranging the objects subjected to the global displacement mapping and the regional displacement mapping in order of distance from a viewpoint;
A tangent space correction step in which the height direction of the tangent space between adjacent polygons is the same in the global displacement mapping step;
A double displacement mapping method comprising: The regional displacement mapping step includes:
Searching for an intersection between the surface of the protruding object defined by the regional displacement information and the observer's line of sight, and displaying the shape of the protruding object on the polygon surface; and
Adding an additional polygon surface, and displaying the shape of the protruding object corresponding to the excess exceeding the boundary of the polygon surface on the additional polygon surface; and
A steep slope detection step of detecting a steep slope in the shape of the protruding object;
A steep slope correction step for expressing an additional texture on the steep slope;
The double displacement mapping method according to claim 12, comprising: The intersection search step includes:
A sequential search step for searching for an approximate position of an intersection between the surface of the object defined by the regional displacement information and the line of sight;
A binary search step for searching for an exact intersection point and a height of the intersection point from a rough position of the intersection point;
The double displacement mapping method according to claim 13, comprising: The sequential search step includes:
Dividing between the upper surface of the virtual volume constituted by the regional displacement information and the polygon surface at regular intervals,
Search for a divided intersection where the upper surface of the virtual volume and the divided surface and the line of sight intersect,
Starting with the uppermost divided intersection, the height value of the divided surface is compared with the height value stored in the displacement information, and the height value of the divided surface is stored in the displacement information. If the height value is higher than the height value stored in the displacement information, the inspection is continued if the height value of the divided surface is lower than the height value stored in the displacement information.
The double displacement mapping method according to claim 14, wherein an approximate position of the intersection is set between the divided intersection where the inspection is stopped and the divided intersection on the intersection.   In the excess display step, as the additional polygon surface, the polygon surface is expanded on the same surface, and the shape of the protruding object corresponding to the excess is displayed on the expanded polygon surface. The double displacement mapping method according to claim 13.   In the excess display step, a vertical plane of the polygon surface is added as the additional polygon plane adjacent to a boundary of the polygon surface where an excess occurs depending on the direction of the line of sight, and the added vertical plane is 14. The double displacement mapping method according to claim 13, wherein the shape of the protruding object corresponding to an excess is displayed.   The double displacement mapping method according to claim 17, wherein the excess display step maps the tangent space between the polygon surface and the vertical surface. The excess display step includes:
The spatial coordinates of the intersection C between the line of sight and the vertical plane are (dx, dy), and the spatial coordinates of the intersection A between the virtual extension plane obtained by extending the polygon plane on the same plane and the line of sight are (x ′, y '), The point C is mapped to the point A using the following formula:
Here, (vx, vy, vz) is a value in the tangent space coordinate system of the polygon surface, (dx, dy) is a value in the uv space coordinate system of the vertical surface, and (x ′, y ′) 18. The double displacement mapping method according to claim 17, wherein is a value of a spatial coordinate system of the virtual extension plane, and the shape of the protruding object is displayed using the texture coordinates of the point A.   The double displacement mapping method according to claim 13, wherein, in the excess display step, only the shape of the protruding object corresponding to the excess is rendered without displaying the additional polygon surface. . The steeply inclined surface detecting step includes
Dividing between the upper surface of the virtual volume constituted by the regional displacement information and the polygon surface at regular intervals,
Search for a divided intersection where the upper surface of the virtual volume and the divided surface and the line of sight intersect,
Obtaining a height value of an intersection of the surface of the object and the line of sight defined by the regional displacement information, and setting the obtained height value as a first height value;
Obtaining a height value of a divided intersection existing on the upper surface adjacent to the intersection, and setting the obtained height value as a second height value;
The double according to claim 13, wherein when the difference between the first height value and the second height value exceeds a limit value, it is determined that the surface is a steeply inclined surface. Displacement mapping method.   The double displacement mapping method according to claim 13, wherein the steep slope correction step adds a tile-shaped texture to the steep slope.   The double displacement mapping method according to claim 12, wherein the global displacement information is basic terrain information, the regional displacement information is artificial feature information, and the texture information is an aerial photograph. In a method of processing an excess exceeding the boundary of the polygon surface by a computer graphic system that displays the shape of an object protruding on the polygon surface,
Adding a vertical surface of the polygon surface adjacent to the boundary of the polygon surface where the excess occurs depending on the direction of the line of sight, and displaying the protruding object shape corresponding to the excess on the added vertical surface An excess processing method characterized by.   The excess processing method according to claim 24, wherein the tangent space between the polygon surface and the vertical surface is mapped. The spatial coordinates of the intersection C between the line of sight and the vertical plane are (dx, dy), and the spatial coordinates of the intersection A between the virtual extension plane obtained by extending the polygon plane on the same plane and the line of sight are (x ′, y ′). ), The point C is mapped to the point A using the following formula:
Here, (vx, vy, vz) is a value in the tangent space coordinate system of the polygon surface, (dx, dy) is a value in the uv space coordinate system of the vertical surface, and (x ′, y ′) The excess processing method according to claim 24, wherein is a value of a spatial coordinate system of the virtual extension plane, and the shape of the protruding object is displayed using the texture coordinates of the point A.   25. The excess processing method according to claim 24, wherein only the protruding object shape corresponding to the excess is rendered without displaying the added vertical plane.